"This research builds upon a previous study, "Validating Causal Diagrams of Human Health Risks for Spaceflight: An Example Using Bone Data from Rodents," conducted last year by scientists including Robert Reynolds, Ryan Scott, and Lauren Sanders [11]. The primary objective of our experiment was to explore the feasibility of leveraging pre-existing Bayesian structure learning algorithms to replicate the causal diagrams generated in the aforementioned study. The focus of the original research was the investigation of Human Health Risks for Spaceflight, specifically pertaining to the causes of diminished bone strength due to gravitational changes. To achieve this, datasets from four distinct experiments — Dubeé (2014), Keune (2015), Keune (2016), Ko (2020 pt. 1), and Ko (2020 pt. 2) — were utilized. However, as the research progressed, the emphasis shifted from merely replicating these specific Directed Acyclic Graphs (DAGs) to a broader inquiry: the potential of creating Bayesian network structures that accurately depict causal relationships validated by domain experts. This journey revealed challenges stemming from data quality and the inherent limitations of working with small datasets. As a result, the study concludes by acknowledging the current constraints in obtaining algorithms to perform optimally with small datasets and suggests avenues for future research. These include exploring larger datasets and refining algorithm parameterization methods to enhance the learning of causal relations."
- Sierra Martin, "Constraint-Based Algorithms Used to Recreate Expert Causal Diagrams: HITON-PC"
The MIT License (MIT)
Copyright (c) 2023 Sierra Martin
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.